Optimization of Compositions of MgO-B2O3-SiO2 Slags Using Artificial Neural Networks and Genetic Alsorithm

The relation between the compositions of MgO-B 2 O 3 -SiO 2 slag and the efficiencies of extraction of boron has been modelled and optimal composition has been predicted using artificial neural networks (ANN). This is the first study where ANN and genetic algorithms have been combined for optimizing a process of extractive metallurgy. The results show that it is an effective way to optimize complicated processes with unknown mechanisms.